wehub-resource-sync
122 lines
4.0 KiB
Python
122 lines
4.0 KiB
Python
# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at\
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import unittest
|
|
|
|
import numpy as np
|
|
from get_program import (
|
|
get_dummy_program,
|
|
)
|
|
|
|
from paddle.tensorrt.export import (
|
|
Input,
|
|
PrecisionMode,
|
|
TensorRTConfig,
|
|
convert_to_trt,
|
|
)
|
|
from paddle.tensorrt.util import (
|
|
predict_program,
|
|
)
|
|
|
|
|
|
class TestConverterDummy(unittest.TestCase):
|
|
def test_paddle_to_tensorrt_conversion_dummy(self):
|
|
program, scope, param_dict = get_dummy_program()
|
|
|
|
# Set input
|
|
input_config = Input(
|
|
min_input_shape=(1, 64),
|
|
optim_input_shape=(4, 64),
|
|
max_input_shape=(8, 64),
|
|
input_data_type='float32',
|
|
)
|
|
_, input_optim_data, _ = input_config.generate_input_data()
|
|
# Create a TensorRTConfig with inputs as a required field.
|
|
trt_config = TensorRTConfig(inputs=[input_config])
|
|
trt_config.precision_mode = PrecisionMode.FP16
|
|
trt_config.ops_run_float = "pd_op.add"
|
|
trt_config.optimization_level = 5
|
|
trt_config.disable_passes = ['dead_code_elimination_pass']
|
|
|
|
output_var = program.list_vars()[-1]
|
|
|
|
# get original results(for tests only)
|
|
output_expected = predict_program(
|
|
program, {"input": input_optim_data}, [output_var]
|
|
)
|
|
# get tensorrt_engine_op(converted_program)
|
|
program_with_trt = convert_to_trt(program, trt_config, scope)
|
|
output_var = program_with_trt.list_vars()[-1]
|
|
|
|
# run inference(converted_program)
|
|
output_converted = predict_program(
|
|
program_with_trt, {"input": input_optim_data}, [output_var]
|
|
)
|
|
|
|
# Check that the results are close to each other within a tolerance of 1e-2
|
|
np.testing.assert_allclose(
|
|
output_expected[0],
|
|
output_converted[0],
|
|
rtol=1e-2,
|
|
atol=1e-2,
|
|
err_msg="Outputs are not within the 1e-2 tolerance",
|
|
)
|
|
|
|
def test_paddle_to_tensorrt_collect_shape(self):
|
|
program, scope, param_dict = get_dummy_program()
|
|
|
|
# Set input
|
|
input_data = tuple(
|
|
np.random.rand(n, 64).astype(np.float32) for n in (1, 4, 8)
|
|
)
|
|
input_optim_data = input_data[1]
|
|
input_config = Input(warmup_data=input_data)
|
|
|
|
# Create a TensorRTConfig with inputs as a required field.
|
|
trt_config = TensorRTConfig(inputs=[input_config])
|
|
trt_config.precision_mode = PrecisionMode.FP16
|
|
trt_config.ops_run_float = "pd_op.add"
|
|
trt_config.optimization_level = 5
|
|
trt_config.disable_passes = ['dead_code_elimination_pass']
|
|
|
|
# get tensorrt_engine_op(converted_program)
|
|
program_with_trt = convert_to_trt(program, trt_config, scope)
|
|
|
|
output_var = program.list_vars()[-1]
|
|
|
|
# get original results(for tests only)
|
|
output_expected = predict_program(
|
|
program, {"input": input_optim_data}, [output_var]
|
|
)
|
|
|
|
output_var = program_with_trt.list_vars()[-1]
|
|
|
|
# run inference(converted_program)
|
|
output_converted = predict_program(
|
|
program_with_trt, {"input": input_optim_data}, [output_var]
|
|
)
|
|
|
|
# Check that the results are close to each other within a tolerance of 1e-2
|
|
np.testing.assert_allclose(
|
|
output_expected[0],
|
|
output_converted[0],
|
|
rtol=1e-2,
|
|
atol=1e-2,
|
|
err_msg="Auto shape collection outputs mismatch",
|
|
)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|